Investment Banking Analyst Interview Questions

I start by forecasting unlevered free cash flow for five to ten years: revenue build, operating margin assumptions, taxes, capex, and changes in working capital. I calculate the discount rate using WACC, weighting the cost of equity from CAPM and the after-tax cost of debt by their respective proportions in the capital structure. I discount each year's projected free cash flow back to present value, then add the terminal value, usually built with a perpetuity growth rate of 2 to 3% or an exit multiple pulled from comparable transactions. I sum the present values to get enterprise value, then bridge to equity value by subtracting net debt and minority interest. I always sensitise the output on WACC and terminal growth in a two-way data table, because those two inputs typically swing the valuation by 20 to 30% on their own. A DCF without a sensitivity table is just one person's opinion dressed up as maths.

Comps are fast, market-based, and hard to argue with because they reflect what investors are actually paying today for similar businesses. The weakness is that the whole market can be mispriced, and comps assume your target company is truly similar to the peer set, which is rarely perfectly true: growth rates, margins, and capital intensity all need adjusting for. A DCF is intrinsic, it doesn't depend on market sentiment, and it forces you to state your assumptions explicitly rather than hiding them inside a multiple. Its weakness is that it's extremely sensitive to long-dated assumptions nobody can forecast with real precision, especially terminal value. In practice I never present one in isolation. I build a football field chart with DCF, trading comps, precedent transactions, and an LBO implied value where relevant, then explain to the client why the ranges converge or diverge.

A sponsor acquires a company using a mix of equity and a large proportion of debt, often 50 to 70% of the purchase price, secured against the target's own assets and cash flows. Returns come from three levers. First, EBITDA growth: increasing revenue or margins over the hold period. Second, multiple expansion: exiting at a higher EBITDA multiple than the entry multiple, though sponsors typically underwrite deals assuming flat or slightly lower exit multiples to be conservative. Third, and often the largest driver in a well-structured deal, deleveraging: using the company's own cash flow to pay down debt over the hold period, which increases the equity value directly since enterprise value stays constant while debt shrinks. I build a simple LBO by projecting the entry capital structure, running the operating case, tracking the debt paydown schedule against a cash flow sweep, then calculating the exit equity value and back-solving for IRR and money multiple.

A traditional DCF still works technically, but the terminal value assumption carries almost all the weight because near-term cash flows are negative or negligible, so you need to be honest that you're really valuing a story about the future, not the present. I'd lean more heavily on revenue multiples against comparable companies at a similar growth stage, adjusted for growth rate and gross margin differences, since EBITDA or earnings multiples don't work when earnings are negative. I'd also build out a path-to-profitability model explicitly: at what revenue scale does the business cross into positive operating margin, and how sensitive is that to unit economics like customer acquisition cost and lifetime value. For very early-stage or venture-style situations I'd also look at recent funding rounds and post-money valuations as a market check. The honest answer to give a client is that any single number here should come with a wide range and clearly stated assumptions, not false precision.

I worked on a sell-side process for a mid-market industrial business with roughly £160 million in revenue. My role as the junior analyst was building and maintaining the operating model, running the comparable company and precedent transaction screens, and producing the first drafts of the management presentation and teaser. The most demanding part was a late change: three weeks before first-round bids, the client updated their FY guidance downward by about 8%, which meant rebuilding the entire model, re-running every valuation output, and updating 40-plus pages of the CIM overnight so the associate and VP could review before the bank opened the next morning. I built a checklist to track every page and exhibit that referenced the old numbers so nothing got missed in the rush. We hit the deadline with the materials fully consistent, and the process ultimately closed with four bidders in the first round.

During a live deal I was staffed on two live processes simultaneously for about six weeks, both requiring same-week deliverables. I was regularly working past 2am and the volume of model updates, formatting requests, and diligence requests was more than I could fully absorb without a system. I started triaging every incoming request by deadline and by who was asking, since a same-day request from the VP on a live process took priority over a formatting tweak that could wait until morning. I also built templates for recurring outputs, like the weekly diligence tracker, so I wasn't rebuilding the same document from scratch each time. I flagged to my staffer early when I really couldn't hit two deadlines at once rather than silently missing one, which let them reprioritise or pull in a second-year analyst for backup. Both deals closed on schedule. I learned that visibility into capacity constraints, given early, is more valuable to the team than quietly trying to do everything myself.

Two nights before a board presentation, I noticed the debt schedule in our model was pulling the prior year's interest rate assumption instead of the current forward curve we'd updated that week, which understated projected interest expense by roughly £350,000 a year across the forecast period. I traced it to a broken link from when the tab had been reorganised. I flagged it immediately to the associate rather than quietly fixing it and moving on, since the output had already been used in a page that had gone out in a preliminary draft. We corrected the model, re-ran the sensitivities, and I built a version-control note into the file going forward so any structural change to the debt tab got flagged to the whole team. On the flip side, I've also had a formatting error in a comp set caught by my VP during a late-night review, and the response that mattered most wasn't apologising repeatedly, it was fixing it fast and building a check into my process so it didn't recur.

Net income from the bottom of the income statement flows into retained earnings on the balance sheet and is the starting line of the cash flow statement. On the cash flow statement, the operating section adjusts net income for non-cash items like depreciation and amortisation, which come from the balance sheet's fixed asset schedule, and for changes in working capital accounts like receivables, payables, and inventory. The investing section captures capex, which reduces cash and increases gross PP&E on the balance sheet, with depreciation subsequently reducing net PP&E over time. The financing section captures debt issuance or repayment and dividends or share buybacks, which flow into the balance sheet's debt and equity lines. The ending cash balance from the cash flow statement becomes the cash line on the balance sheet, and the balance sheet balances because assets equal liabilities plus equity by construction, assuming every change has been captured correctly across all three statements. If a model doesn't balance, the error is almost always a missing or double-counted link between these three.

On the income statement, EBIT decreases by £10, and assuming a 25% tax rate, net income falls by £7.50. On the cash flow statement, you start with the lower net income of negative £7.50, then add back the £10 of depreciation since it's a non-cash expense, giving a net increase to cash flow from operations of £2.50, which is just the tax shield on the depreciation. On the balance sheet, cash increases by £2.50 from that improved operating cash flow. PP&E decreases by £10 due to the higher depreciation. Retained earnings decreases by £7.50, matching the lower net income. Checking the balance: assets are down £7.50 in total, which is £2.50 up in cash minus £10 down in PP&E, and that matches the £7.50 decrease in retained earnings on the liabilities and equity side. The balance sheet still balances.

I'd lead with the strategic rationale before any numbers: why this specific target makes sense for this specific buyer right now, whether that's filling a product gap, gaining share in a geography, or removing a competitor. I'd size the market opportunity and quantify potential synergies conservatively, splitting cost synergies, which are easier to underwrite and defend, from revenue synergies, which deal teams and boards are rightly sceptical of. I'd show an illustrative accretion or dilution analysis on EPS, since that's often the first question a board asks, and a preliminary view on financing structure, cash, stock, or debt, and how that shifts the accretion maths. I'd flag the two or three biggest risks upfront rather than waiting for the client to find them, whether that's regulatory approval, integration complexity, or valuation gap with the target's own expectations. An MD reviewing this wants to see that I've anticipated the client's hardest questions before I walk into the room, not just that I built a clean model.